Ant Group open-sources Ming-3.0-tts speech models under MIT license
Overview
Ant Group's inclusionAI team has open-sourced Ming-3.0-tts, a speech-generation family with a 0.5B dense model and a 7.9B-total, 1.3B-active MoE tiny model, released under the MIT license.
The models offer voice cloning, voice design from text, over 100 Chinese and five English built-in voices, and control over emotion, speed, pitch and volume. Code, a task cookbook and evaluation tables are on GitHub, with weights on Hugging Face and ModelScope.
The project's README lists weak spots, including lower pitch-control accuracy and a weaker Korean error rate for the tiny model. The project asks users not to clone real people's voices without consent and to label synthetic speech where it could be mistaken for a real recording.
Written by AI from the articles below · updated Oct 19, 10:25 PM ET
Check the sources:
Article timeline
The articles in this story. Times are ET.
- PandailyNewsAnt Group open-sources Ming-3.0-tts speech models with voice design and fine control
AIAnt Group's inclusionAI team has open-sourced Ming-3.0-tts under the MIT license, with a 0.5B dense model and a 7.9B-total, 1.3B-active MoE tiny model. The models offer voice cloning, voice design from text, over 100 Chinese and five English built-in voices, and fine control over emotion, speed, pitch and volume. The README lists weak spots, including lower pitch-control accuracy and a weaker Korean error rate for the tiny model.
Heat Trend
Not enough continuous observations to show a trend yet.